U.S. patent application number 10/935833 was filed with the patent office on 2006-03-09 for position location signaling method apparatus and system utilizing orthogonal frequency division multiplexing.
Invention is credited to Norman F. Krasner.
Application Number | 20060050625 10/935833 |
Document ID | / |
Family ID | 35506615 |
Filed Date | 2006-03-09 |
United States Patent
Application |
20060050625 |
Kind Code |
A1 |
Krasner; Norman F. |
March 9, 2006 |
Position location signaling method apparatus and system utilizing
orthogonal frequency division multiplexing
Abstract
Position location signaling system, apparatus, and method are
disclosed. Position location beacons can each be configured to
transmit a frequency interlaced subset of orthogonal frequencies
spanning substantially an entire channel bandwidth. The orthogonal
frequencies can be pseudorandomly or uniformly spaced, and each
beacon can be allocated an equal number of orthogonal frequencies.
Each frequency of the interlaced subset of orthogonal frequencies
can be modulated with an element of a predetermined data sequence.
A mobile device can receive one or more of the beacon signals and
determine a position using a position location algorithm that
determines position in part on an arrival time of the beacon
signal. Where the mobile device can receive three or more beacon
signals, the mobile device can perform position location by
trilateration to the beacon positions based, for example, on a time
difference of arrival.
Inventors: |
Krasner; Norman F.; (Emerald
Hills, CA) |
Correspondence
Address: |
QUALCOMM, INC
5775 MOREHOUSE DR.
SAN DIEGO
CA
92121
US
|
Family ID: |
35506615 |
Appl. No.: |
10/935833 |
Filed: |
September 7, 2004 |
Current U.S.
Class: |
370/208 |
Current CPC
Class: |
H04L 27/2601 20130101;
G01S 1/042 20130101; G01S 1/20 20130101; G01S 1/045 20130101; G01S
1/04 20130101; G01S 5/10 20130101 |
Class at
Publication: |
370/208 |
International
Class: |
H04J 11/00 20060101
H04J011/00 |
Claims
1. A method of transmitting position location signals, the method
comprising: defining a plurality (Q) of orthogonal frequencies,
wherein each of the Q orthogonal frequencies are separated from one
another by a multiple of a fixed frequency spacing (w); selecting a
first subset of orthogonal frequencies from the plurality of
orthogonal frequencies; generating a first Orthogonal Frequency
Division Multiplex (OFDM) symbol based on the first subset of
orthogonal frequencies and substantially orthogonal to a second
OFDM symbol generated from a second subset of orthogonal
frequencies from the plurality of orthogonal frequencies; and
transmitting a position location signal containing said first OFDM
symbol from a first geographical location distinct from a second
geographical location from which the second OFDM symbol is
transmitted.
2. The method of claim 1, wherein defining the plurality of
orthogonal frequencies comprises defining a plurality of uniformly
spaced orthogonal frequencies.
3. The method of claim 1, wherein defining the plurality of
orthogonal frequencies comprises defining the plurality of
orthogonal frequencies spanning substantially an entire channel
bandwidth.
4. The method of claim 1, wherein defining the plurality of
orthogonal frequencies comprises defining the plurality of
orthogonal frequencies spanning substantially 5.5 MHz.
5. The method of claim 1, wherein selecting the first subset of
orthogonal frequencies comprises selecting a subset of uniformly
spaced orthogonal frequencies from the plurality of orthogonal
frequencies.
6. The method of claim 1, wherein selecting the first subset of
orthogonal frequencies comprises selecting a subset of randomly
spaced orthogonal frequencies from the plurality of orthogonal
frequencies.
7. The method of claim 1, wherein selecting the first subset of
orthogonal frequencies comprises selecting a subset of
pseudorandomly spaced orthogonal frequencies from the plurality of
orthogonal frequencies.
8. The method of claim 1, wherein selecting the first subset of
orthogonal frequencies comprises: defining a number (M) of distinct
position location signals; and selecting Q/M of orthogonal
frequencies from the plurality of orthogonal frequencies.
9. The method of claim 1, further comprising modulating each
frequency in the first subset of orthogonal frequencies.
10. The method of claim 1, further comprising: generating a binary
data sequence; and modulating each frequency in the first subset of
orthogonal frequencies based in part on the binary sequence.
11. The method of claim 10, wherein generating the binary sequence
comprises generating a pseudorandom binary sequence.
12. The method of claim 10, wherein generating the binary sequence
comprises generating a pseudo random binary sequence of a type
selected from a group comprising a maximal length binary sequence,
a Barker code, a Gold code, and a Walsh code.
13. The method of claim 10, wherein modulating each frequency in
the subset of orthogonal frequencies comprises phase modulating
each frequency in the subset of orthogonal frequencies.
14. The method of claim 1, further comprising: generating a
nonbinary signal; and modulating each frequency in the first subset
of orthogonal frequencies based in part on the nonbinary
signal.
15. The method of claim 14, wherein modulating each frequency
comprises modulating one of a phase, amplitude, or combination of
phase and amplitude.
16. The method of claim 14, wherein the nonbinary signal comprises
a constant signal over a period at least 1/w.
17. The method of claim 1 wherein transmitting the signal
containing said first OFDM symbol comprises transmitting a first
signal time synchronized to a first external event.
18. The method of claim 17, wherein transmitting the signal
containing said second OFDM symbol comprises transmitting a second
signal time synchronized to a second external event.
19. The method of claim 1, wherein generating the first OFDM symbol
comprises transforming the first subset of orthogonal frequencies
to a time domain signal using an Inverse Fourier Transform.
20. The method of claim 1, wherein generating the first OFDM symbol
comprises transforming the first subset of orthogonal frequencies
to a time domain signal using a Q-point Inverse FFT.
21. The method of claim 1, wherein transmitting the signal
containing the first OFDM symbol comprises wirelessly transmitting
an electromagnetic signal containing said first OFDM symbol.
22. The method of claim 1, wherein transmitting the signal
containing the first OFDM symbol comprises periodically
transmitting the OFDM symbol.
23. The method of claim 1, wherein transmitting the signal
containing the first OFDM symbol comprises periodically
transmitting the signal containing the first OFDM symbol using a
television broadcast transmitter.
24. The method of claim 1, wherein a duration of said first OFDM
symbol is substantially equal to 1/w.
25. The method of claim 1, wherein a duration of said first OFDM
symbol is extended to be greater than 1/w.
26. A method of transmitting position location signals, the method
comprising: defining a plurality (Q) of orthogonal frequencies;
defining a number (M) of distinct position location signals;
generating M disjoint subsets of orthogonal frequencies from the
plurality of orthogonal frequencies; generating at least two
Orthogonal Frequency Division Multiplex (OFDM) symbols
corresponding to at least two of the M subsets; and transmitting
one of the at least two OFDM symbols periodically from a first
geographical location over a wireless communication channel, the
first geographical location distinct from a second geographical
location from which the second of the at least two OFDM symbols is
transmitted.
27. The method of claim 26, wherein generating M disjoint subsets
of orthogonal frequencies comprises generating M subsets having Q/M
mutually exclusive orthogonal frequencies from the plurality of
orthogonal frequencies.
28. The method of claim 26, further comprising Phase Shift Key
modulating at least one of the M disjoint subsets of orthogonal
frequencies using a pseudorandom binary sequence.
29. A method of transmitting position location signals, the method
comprising: generating a first frequency interlaced Orthogonal
Frequency Division Multiplex (OFDM) signal from a first subset of
orthogonal frequencies; synchronizing the first frequency
interlaced OFDM signal to a time reference; generating a second
frequency interlaced OFDM signal from a subset of orthogonal
frequencies disjoint from said first subset; synchronizing the
second frequency interlaced OFDM signal to the time reference; and
wirelessly transmitting the first and the second frequency
interlaced OFDM signals from two different geographical
locations.
30. A position location signal generating apparatus, the apparatus
comprising: means for generating at least two out of a total of M
subsets of orthogonal frequencies from a plurality of Q orthogonal
frequencies; means for generating at least two Orthogonal Frequency
Division Multiplex (OFDM) symbols, each of said symbols constructed
from a different subset from the M subsets; means for transmitting
one of said OFDM symbols periodically over a first wireless
communications link; and means for transmitting a second of said
OFDM symbols periodically over a second wireless communications
link.
31. A position location signal generating apparatus, the apparatus
comprising: means for generating at least two frequency interlaced
Orthogonal Frequency Division Multiplex (OFDM) signals from a
subset of orthogonal frequencies; means for synchronizing each of
said frequency interlaced OFDM signals to a time reference; and
means for concurrently transmitting each of said frequency
interlaced OFDM signals from each of two different geographical
locations.
32. A position location signal generating apparatus, the apparatus
comprising: an orthogonal signal generator configured to generate
at least a subset of orthogonal carriers out of a larger set of Q
orthogonal frequency carriers; a modulation data module configured
to generate a pseudorandom data sequence; a modulator coupled to
the modulation data module and configured to modulate the subset of
orthogonal carriers based in part on a pseudorandom data sequence;
an Orthogonal Frequency Division Multiplex (OFDM) modulator having
an input coupled to the orthogonal signal generator, and configured
to generate a first OFDM symbol based in part on the subset of
orthogonal carriers; and a transmitter coupled to the OFDM
modulator and configured to wirelessly transmit the first OFDM
symbol from a first geographical location distinct from a second
geographical location from which a second OFDM symbol, orthogonal
to the first OFDM symbol, is transmitted.
33. The apparatus of claim 32, wherein the pseudorandom data
sequence comprises a code selected from the group comprising a Gold
code, a Barker code, a maximal length code, and a Walsh code.
34. The apparatus of claim 32, wherein the pseudorandom data
sequence comprises a sequence of a length equal to a number of
carriers in the subset of orthogonal frequencies.
35. The apparatus of claim 32, wherein the modulator is configured
to perform modulation on each of the carriers of the subset of
orthogonal carriers based in part on the pseudorandom data
sequence, wherein said modulation is by one of phase modulation,
amplitude modulation, or combined phase and amplitude
modulation.
36. The apparatus of claim 32, wherein the modulator is configured
to Binary Phase Shift Key (BPSK) modulate each of the carriers of
the subset of orthogonal carriers based on a value of a
corresponding bit in the binary sequence.
37. The apparatus of claim 32, wherein the OFDM modulator
comprises: an Inverse Fast Fourier Transform (IFFT) module
configured to perform an IFFT of the subset of orthogonal carriers
of size at least Q; and a parallel to serial converter coupled to
the output of the IFFT module and configured to generate a serial
output from an output of the IFFT module.
38. The apparatus of claim 32, wherein the orthogonal signal
generator is configured to generate one of M subsets of orthogonal
carriers by generating Q/M orthogonal carriers.
39. The apparatus of claim 38, wherein the Q/M orthogonal carriers
comprise uniformly spaced orthogonal carriers.
40. The apparatus of claim 38, wherein the Q/M orthogonal carriers
comprise pseudo randomly spaced orthogonal carriers.
41. A method of position location, the method comprising: receiving
an Orthogonal Frequency Division Multiplex (OFDM) signal;
performing a cross-correlation operation of said received OFDM
signal against a reference signal; determining a presence of a
received OFDM symbol from said cross-correlation operation; and
determining a location based at least in part on the OFDM
symbol.
42. The method of claim 41, further comprising determining a
Doppler shift of the OFDM symbol.
43. The method of claim 42, wherein determining the Doppler shift
comprises: hypothesizing multiple Doppler shifts; determining a
corresponding strength of said received OFDM symbol in response to
each of said hypothesized Doppler shifts; and performing an
interpolation operation upon said corresponding strengths.
44. The method of claim 41, further comprising: rejecting spurious
cross correlation products within the received OFDM signal by
hypothesizing multiple Doppler shifts; determining corresponding
strengths of said cross-correlation in response to said
hypothesized Doppler shifts; and examining a variation in said
strengths.
45. The method of claim 44, wherein rejecting cross correlation
products comprises rejecting the OFDM symbol if the variation in
said strengths differs by more than a predetermined threshold.
46. The method of claim 41, wherein receiving the OFDM signal
comprises receiving a frequency interlaced OFDM symbol generated
from a subset of orthogonal frequencies spanning substantially a
channel bandwidth of a broadcast band.
47. The method of claim 41, wherein receiving the OFDM signal
comprises receiving at least one of M frequency interlaced OFDM
symbols, each of the M frequency interlaced OFDM symbols comprising
a subset of Q/M frequencies from a set of Q orthogonal
frequencies.
48. The method of claim 41, wherein receiving the OFDM signal
comprises receiving a frequency interlaced OFDM symbol from a
television broadcast transmitter.
49. The method of claim 41, wherein determining the location
comprises: determining a pseudorange based on said
cross-correlation operation; and determining the location based in
part on the pseudorange.
50. The method of claim 49, wherein determining the location
further comprises transmitting the pseudorange to an entity
distinct from a system transmitting the OFDM signal.
51. The method of claim 41, wherein performing the
cross-correlation comprises performing one of a matched filter
operation or an FFT operation.
52. A mobile device configured for position location, the device
comprising: means for receiving an Orthogonal Frequency Division
Multiplex (OFDM) signal; means for cross-correlating at least a
portion of the OFDM signal to one of a predetermined number of OFDM
symbols; and means for determining a location based at least in
part on a result of said cross correlation operation.
53. A mobile device configured for position location, the device
comprising: a receiver configured to receive an Orthogonal
Frequency Division Multiplex (OFDM) signal; a correlator configured
to correlate the received OFDM signal against a plurality of
frequency interlaced OFDM symbols; and a position location module
configured to determine a location based in part on an output of
the correlator if at least one frequency interlaced OFDM symbol
correlates with the received OFDM signal by greater than a
predetermined correlation threshold.
54. The device of claim 53, wherein the correlator comprises: a
matched filter module configured to provide matched filter response
for each OFDM symbol; and a peak detection module coupled to an
output of the matched filter module.
55. The device of claim 53, wherein the correlator comprises: a
Fast Fourier Transform (FFT) module configured to transform the
received OFDM signal to frequency domain signal; a modulator
configured to modulate the frequency domain signal; an IFFT module
configured to transform an output of the modulator to a time domain
signal; and a peak detection module coupled to the output of the
IFFT producing an indication of said level of correlation.
Description
FIELD
[0001] The disclosure relates to the field of position location.
More particularly, the disclosure relates to wireless position
location systems, signaling, and devices.
BACKGROUND
[0002] In many applications it may be advantageous to have the
ability to determine a position of a mobile device. Position
location may be helpful for navigation, tracking, or orientation
applications. The continual advancement of the performance of
portable electronics, particularly the advancements in the
performance of processors, allows position location capabilities to
be added in a variety of devices.
[0003] For example, it may be desirable for an operator of a mobile
telecommunications system such as a cellular telecommunications
system to be able to determine the position of a mobile handset
during communication with a base transceiver station (BTS) of the
system. A system operator may desire position location
capabilities, for example, to satisfy the U.S. Federal
Communications Commission (FCC) E911 emergency position location
mandate.
[0004] Mobile devices may implement one or more position location
techniques depending on the position location signaling methods
used in the position location system. For example, a mobile device
may use time of arrival (TOA), time difference of arrival (TDOA),
advanced forward link trilateration (AFLT) or some other position
location technique. Examples of position location systems include
those that are based on the Global Positioning System (GPS), those
that augment the GPS system with terrestrial based beacons such as
Assisted GPS systems, and terrestrial based beacon position
location systems.
[0005] Most terrestrial ranging systems incorporate a pseudo noise
(PN) code in a direct sequence spread spectrum configuration. A
mobile device can identify a particular source, in part, by
correlating a received PN spread signal with an internally
generated version. Unfortunately, PN codes typically exhibit modest
cross correlation properties unless very long PN codes are used.
However, the use of long PN codes increase the complexity,
bandwidth, or time required to obtain a position location fix.
Additionally, because a mobile station in a terrestrial based
system can receive widely disparate signal powers, even relatively
low cross correlation properties can interfere with the mobile
station's ability to identify signal sources.
[0006] Therefore, it is desirable to have a position location
signaling technique, system, and device that allow for high
performance position location in a variety of conditions and yet
may be implemented in a practical manner.
SUMMARY
[0007] Position location signaling system, apparatus, and method
are disclosed. Position location beacons can be configured to
transmit an interlaced subset of orthogonal frequencies spanning
substantially an entire channel bandwidth. The subset of orthogonal
frequencies are preferentially pseudorandomly spaced; however,
uniform spacing may be used. Each beacon can be allocated an equal
number of orthogonal frequencies. Adjacent beacons can be assigned
mutually exclusive subsets of orthogonal frequencies. Each
frequency of the interlaced subset of orthogonal frequencies can be
modulated with an element of a predetermined data sequence. A
mobile device can receive one or more of the beacon signals and
determine a position using a position location algorithm that
determines position in part on an arrival of the beacon signal.
Where the mobile device can receive three or more beacon signals,
the mobile device can perform position location by trilateration to
the beacon positions based, for example, on pseudoranges or a time
difference of arrival.
[0008] One aspect includes a method of generating position location
signals. The method includes defining a plurality (Q) of orthogonal
frequencies, selecting a subset of orthogonal frequencies from the
plurality of orthogonal frequencies, generating an Orthogonal
Frequency Division Multiplex (OFDM) symbol based on the subset of
orthogonal frequencies, and transmitting the OFDM symbol.
[0009] Another aspect includes a method of generating position
location signals. The method includes defining a plurality (Q) of
orthogonal frequencies, defining a number (M) of distinct position
location signals, generating M subsets of orthogonal frequencies
from the plurality of orthogonal frequencies, generating an OFDM
symbol corresponding to at least one of the M subsets, and
transmitting the OFDM symbol periodically over a wireless
communication system.
[0010] Yet another aspect includes a method of generating position
location signals. The method includes generating a frequency
interlaced OFDM signal from a subset of orthogonal frequencies,
repeating at least a portion of the OFDM signal at least once to
generate a redundant OFDM signal ("cyclically extending" the
signal), synchronizing the redundant OFDM signal to a time
reference, and wirelessly transmitting the redundant OFDM
signal.
[0011] Yet another aspect includes a position location signal
generating apparatus. The apparatus includes means for generating
at least one of M subsets of orthogonal frequencies from a
plurality of Q orthogonal frequencies, means for generating an OFDM
symbol corresponding to at least one of the M subsets, and means
for transmitting the OFDM symbol periodically over a wireless
communication system.
[0012] Yet another aspect includes a position location signal
generating apparatus. The apparatus includes means for generating a
frequency interlaced OFDM signal from a subset of orthogonal
frequencies, means for synchronizing the OFDM signal to a time
reference, and means for wirelessly transmitting the OFDM
signal.
[0013] Yet another aspect includes a position location signal
generating apparatus. The apparatus includes an orthogonal signal
generator configured to generate at least a subset of orthogonal
carriers out of a larger set of Q orthogonal frequency carriers, an
OFDM modulator having an input coupled to the orthogonal signal
generator, and configured to generate an OFDM symbol based in part
on the subset of orthogonal carriers, and a transmitter coupled to
the OFDM modulator and configured to wirelessly transmit the OFDM
symbol.
[0014] Yet another aspect includes a method of position location.
The method includes receiving an OFDM signal, determining a
received OFDM symbol from the OFDM signal, and determining a
location based at least in part on the OFDM symbol.
[0015] Yet another aspect includes a mobile device configured for
position location. The device includes a receiver configured to
receive an OFDM signal, a correlator configured to correlate the
received OFDM signal with one of a plurality of frequency
interlaced OFDM symbols, and a position location module configured
to determine a location based in part on an output of the
correlator if at least one frequency interlaced OFDM symbol
correlates with the received OFDM signal.
[0016] In another aspect an OFDM position location symbol, or a
multiplicity of such symbols, are transmitted in a time multiplexed
manner with a OFDM communication signal or signals. Periodically,
or on demand, the OFDM communication signal is interrupted and an
OFDM position location signal is substituted.
BRIEF DESCRIPTION OF THE DRAWINGS
[0017] The features, objects, and advantages of the various
embodiments described herein will become more apparent from the
detailed description set forth below when taken in conjunction with
the drawings, in which like elements bear like reference
numerals.
[0018] FIG. 1 is a functional block diagram of an embodiment of a
frequency interlaced OFDM position location system.
[0019] FIG. 2 is a functional diagram of an embodiment of the
frequency spectrum of a pseudorandom interlaced OFDM spectrum.
[0020] FIG. 3 is a functional block diagram of an embodiment of a
frequency interlaced OFDM position location transmitter.
[0021] FIG. 4 is a plot of normalized autocorrelation for a
position location signaling embodiment.
[0022] FIG. 5 is a plot of cross correlation with respect to
frequency offset for a position location signaling embodiment.
[0023] FIGS. 6A-6C are functional block diagrams of embodiments of
frequency interlaced OFDM position location receivers
[0024] FIG. 7 is a flowchart of a method of generating position
location signals.
[0025] FIG. 8 is a flowchart of a method of generating frequency
interlaced OFDM symbols.
[0026] FIG. 9 is a flowchart of a method of position location using
frequency interlaced OFDM symbols.
DETAILED DESCRIPTION
[0027] A position location signaling system, apparatus, and method
based upon an Orthogonal Frequency Division Multiplexing (OFDM)
concept are disclosed. The approach provides improved orthogonality
between simultaneously received position location signals with
large received power differences. The advantages are significant as
compared to alternative configurations such as the use of direct
sequence spread spectrum signaling, utilizing pseudorandom
sequences, such as Gold codes.
[0028] The beacon in the disclosed position location signaling
system preferably uses a wide signal bandwidth to improve a time
resolution achievable at a receiver. Broadband signals also allow
for the use of sophisticated multipath mitigation techniques. A
position location system that is configured to interface with an
existing communication system can be configured to allocate
substantially an entire channel bandwidth or multiple channel
bandwidths for position location signals.
[0029] The position location signals are generated by defining a
set of orthogonal frequencies spanning the available signal
bandwidth. A set of "orthogonal" frequencies is used herein to mean
that the frequency separation between admissible carrier
frequencies of the set are multiples of a frequency difference w,
where w=1/T Hz, where T is the minimum period associated with
symbols constructed as a superposition of carriers having these
frequencies.
[0030] As a further explanation, if two symbols are constructed
from disjoint sets of orthogonal frequencies, with frequency
difference w, then the cross-correlation between the symbols,
integrated over period T=1/w, would ideally be zero. In some cases
the symbol periods of these symbols may be restricted to this
period T. In other cases the symbol period is extended beyond the
period T, by appending to the end of the symbol a portion of the
symbol at its beginning. This extension of the symbol period is
referred to as "cyclic extension." Similarly a portion of the
symbol at the end may be appended to the beginning of the symbol,
again producing a cyclic extension. A combination of the two
methods of appending may be utilized and, alternatively, the symbol
of length T may be repeated a multiplicity of times, thus creating
a very long cyclic extension.
[0031] The period T is referred to as the "basic symbol period" and
the symbol that is extended as the "transmitted symbol period." In
OFDM systems used for communications the period T is referred to as
the "information symbol period." Normally, in this disclosure the
terminology "OFDM symbol" refers to the transmitted symbols-that is
the symbols, including any cyclic extension that may be employed.
Regardless of the length of the transmitted symbol period, a
receiver observing symbols designed to be orthogonal over the
period T, will normally only have zero cross correlation with
respect to one another if the cross correlation is performed over a
period of T or an integer multiple thereof (assuming the symbols
are extended for multiple periods).
[0032] The term "pseudorandom sequence" is used herein to refer to
a sequence of numbers determined by an algorithm whose
characteristics approximate a random sequence of numbers. Examples
of such pseudorandom sequences include maximal length sequences and
Gold code sequences, although many other such sequences exist.
Similarly, the term "pseudorandomly spaced" is used to mean that
the spacing between elements of an series or array is determined
according to some pseudorandom sequence.
[0033] Multiple orthogonal frequency subsets are then defined from
the set of orthogonal frequencies as subsets having disjoint
frequencies relative to one another. Each of the subsets can have
an equal, or at least comparable, number of frequencies. A symbol
constructed from a superposition of carriers with frequencies in a
given orthogonal frequency subset will be orthogonal (over the
period T) to a symbol constructed from carriers from a different
orthogonal subset. Each of the frequencies in a particular subset
of orthogonal frequencies can be modulated or otherwise modified,
as long as the orthogonality is maintained. Typical modulations
used upon each of the frequency carriers include phase shift keying
and QAM modulation, although for position location purposes it is
expected that simple binary or a nonbinary modulation, such as
quaternary phase shift keying would be preferred. Normally the
modulation of the frequencies is held constant during a symbol
period in order to maintain orthogonality.
[0034] In an embodiment, data encoded upon one orthogonal frequency
subset are chosen to be substantially uncorrelated with that of
another orthogonal frequency subset. This approach produces a
two-dimensional orthogonal coding that leads to improved system
performance.
[0035] One subset of orthogonal frequencies can be assigned to a
particular position location beacon. Multiple beacons can be
configured to periodically and concurrently transmit their
respective subset of orthogonal frequencies. In one embodiment, the
beacons can be terrestrial transmitters.
[0036] A mobile device can be configured to receive the
concurrently transmitted position location signals from one or more
of the position location beacons. The mobile device can then
determine its location based in part on the received position
location signals. The mobile device can determine its position, for
example, using a time difference of arrival process or a
trilateration process that determines pseudoranges to the
transmitting beacons. The receiver in the mobile device can
identify the various transmitted position location signal subsets
even in the situation where the position location signal from a
first beacon is received at power levels many orders of magnitude
stronger than the received power of the position location signal
from another beacon. The receiver performance is achievable because
of the low cross correlation properties associated with the
disclosed position location signals.
[0037] FIG. 1 is a functional block diagram of an embodiment of a
frequency interlaced OFDM position location system 100. The
position location system 100 can include a first communication
system having a plurality of position location beacons 120a-120n
configured to transmit corresponding orthogonal multiplexed
position location signals. The position location system 100 may
also include a second communication system having one or more base
stations 130a and 130b. In one embodiment, each of the base
stations 130a and 130b can be coupled to a position location module
140.
[0038] The position location module 140 can include a processor 142
coupled to memory 144 and an almanac 146 configured to store the
locations and other information (for example frequency subset
information) of one or more of the position location beacons
120a-120n. The position location module 140 can be configured to
determine, or assist in determining the position of a mobile
device, such as mobile device 110, operating within the position
location system 100. In some embodiments of the system 100, the
position location module 140 is omitted and the mobile device 110
can determine its location based in part on the received signals
without external assistance.
[0039] In an embodiment of the system 100, the first communication
system can be an existing terrestrial communication system and the
position location signals can be broadcast in addition to the
existing communication signals. Although only three position
location beacons 120a-120n are shown in FIG. 1, any number of
beacons may be included in a system and a mobile device 110 at any
given instant in time may have the ability to receive from all or
less than all of the position location beacons 120a-120n. For
example, the first communication system may be a television
broadcast system, radio broadcast system, or wireless Local Area
Network (LAN) system and existing broadcast antennae can be
configured as position location beacons 120a-120n. In other
embodiments of the system 100, one or more of the position location
beacons 120a-120n can be satellite beacons, aircraft based beacons,
or some other non-terrestrial beacon.
[0040] In an embodiment of the system 100, the second communication
system can be a wireless communication system, such as a wireless
telephone system, and the base stations 130a-130b can be wireless
telephone base stations. The wireless phone system can be, for
example, a Code Division Multiple Access (CDMA) cellular phone
system, a GSM cellular phone system, a Universal Mobile
Telecommunications System (UMTS), or some other system for wireless
communications. Although only two base stations 130a-130b are
shown, any number of stations can be implemented in a particular
communication system.
[0041] The mobile device 110 can be a wireless receiver or a
wireless transceiver and can be, for example, a wireless phone,
cellular phone, cordless phone, radio, position location device,
personal digital assistant, personal communication device, wireless
LAN device, or some other device receiving position location
signals. Because the mobile device 110 can be configured as many
different types of devices, the mobile device 110 may alternatively
be referred to as a mobile station (MS), mobile unit, user
terminal, user device, or portable device.
[0042] In one embodiment, each of the position location beacons
120a-120n can be configured to periodically or continuously
transmit a corresponding interlaced subset of frequencies as
discussed above and further detailed below. A mobile device 110
within the position location system 100 can be configured to
receive the position location signals broadcast by one or more of
the position location beacons 120a-120n. The mobile device 110 can
be configured to correlate or otherwise detect received position
location signals that are above a detection threshold.
[0043] In embodiments where the position location beacons 120a-120n
periodically transmit the position location signals, the position
location beacons 120a-120n should be configured to transmit the
position location signals substantially concurrently in order to
minimize the amount of time that the mobile device 110 needs to
listen for the signals. Position location is simplified if the
transmissions from the position location beacons 120a-120n are
synchronized.
[0044] In one embodiment, it may be desirable to permit positioning
to an accuracy of 30 meters. To achieve this level of accuracy, the
timing errors between the position location beacon 120a-120n
transmissions should be kept below 100 nanoseconds. This, however,
does not take into account any other sources of errors, such as
measurement errors and geometry induced error increases, such as
those attributable to position dilution of precision (PDOP).
Therefore, it may be desirable for the system 100 to maintain the
transmission errors to below 50 nanoseconds relative to one
another. If this is not practical, then of course the ultimate
accuracy will scale accordingly. It may be possible to calibrate
the position location beacon 120a-120n timing errors, if they are
constant, by utilizing the mobile device 110 to determine such
errors, assuming that GPS position location and time-of-day
solutions are available within the MS. Hence, it may be important
to minimize the change in transmission timing, or beacon
synchronization, versus time.
[0045] It may be advantageous for the transmissions from the
position location beacons 120a-120n to contain a data channel that
provides information about the various other position location
beacons 120a-120n present within the vicinity. For example, this
information may be broadcast at a very low rate on some type of
supplementary channel. Information of interest can include the code
numbers and locations of the position location beacons 120a-120n in
an area. The information may include a type of "almanac" and
perhaps other auxiliary information, such as transmission power, or
other factors.
[0046] Complete timing and position location upon beacons 120a-120n
can allow the mobile device 110 to compute its position based upon
this information. If this information is not transmitted by the
position location beacons 120a-120n, it may still possible to
locate the mobile device 110 using information transmitted on
another channel, such as a cellular channel. For example
transmitters 130a and 130b may provide this information. In this
latter case, the cellular channel could send an almanac of position
location beacon 120a-120n information.
[0047] The mobile device 110 may determine its location
independently or may use the assistance of another module, such as
the position location module 140 to determine its position.
Returning ranging information to a network entity or server such as
the position location module 140 can also require messaging. In one
embodiment, the data could be encoded as some type of transmitter
ranging data (for example, so-called AFLT in the CDMA2000 cellular
standard), whose format is already supported in the second
communication system. For example, the format could be in
accordance with IS-801 where the second communication system is a
CDMA communication system. Alternatively, some other type of new
data packet could be transmitted. Current hybrid GPS/AFLT
positioning performed within the IS-801 standard does not directly
support the inclusion of other ranging measurements, such as those
described herein, nor for the transmission of almanac data for
non-cellular base stations in a standardized way. However, such
information may be transmitted on a variety of data channels
supported within such a standard.
[0048] In one embodiment the mobile device 110 can determine a
pseudorange to each of the position location beacons 120a-120n
corresponding to the received position location signals. A
pseudorange is simply a time-of-arrival of a received signal
together with an unknown time bias present in the receiver. The
time bias may be determined as part of the position location
procedure. The mobile device 110 can then determine its location
based in part on the determined pseudoranges.
[0049] In another embodiment, the mobile device 110 can determine a
time difference of arrival based on a pair of received position
location signals, e.g. signals from 120a and 120b. The time
difference of arrival then determines a curve, such as a hyperbola,
on which the mobile device 110 is located. The mobile device 110
can determine another curve based on a different pair of received
position location signals, for example 120b and 120n, and can
determine its location, based in part, on the intersection of the
curves.
[0050] In still another embodiment, the mobile device 110 can
determine a pseudorange to each of the position location beacons
120a-120n corresponding to the received position location signals.
The mobile device 110 can then transmit the pseudorange information
and corresponding position location beacon identification to one or
more base stations, for example 130a, of a second communication
system. The base station 130a can then communicate the pseudorange
and beacon identification information to a position location module
140. The position location module 140 can be configured to
determine the location of the mobile device 110 based in part on
the information provided by the mobile device 110.
[0051] In still other embodiments, the mobile device 110 can
independently determine its location based on some other position
location process. In still other embodiments, the mobile device 110
can determine its location, the position location module 140 can
determine the location of the mobile device 110, or the mobile
device 110 in combination with the position location module 140 can
determine the position of the mobile device 110 using a shared
process.
[0052] FIG. 2 is a functional diagram of an embodiment of a
frequency spectrum of a pseudorandom interlaced orthogonal
frequency spectrum 200. The frequency spectrum 200 shows an overall
signaling bandwidth 210 that can correspond to substantially a
bandwidth of one or more channels in a communication system or can
correspond to a bandwidth allocated for position location
signaling.
[0053] The orthogonal frequencies used for position location
signaling are generated within the overall signaling bandwidth 210.
The signaling frequencies assigned to a given transmitter are shown
as bold horizontal lines to the right of the transmitter numbers.
For example transmitter Tx.sub.4 (211) is assigned frequencies 1,
3, 7, and 12. Although the frequency spectrum 200 in FIG. 2 only
depicts four different signaling frequency subsets corresponding to
four different position location signals, the process can be
generalized to any number M of position location signals.
Initially, Q orthogonal carriers are defined within the overall
signaling bandwidth 210. In the example shown in FIG. 2, the number
of orthogonal carriers Q is equal to sixteen.
[0054] Each of the M position location signals can be constructed
from a subset of the Q total orthogonal carriers. The M position
location signals can be constructed by assigning a subset, q=Q/M,
of carriers to each of the M signals: In one embodiment, q=Q/M is
an integer and all position location signals are an equal number of
carriers. In other embodiments, the number of carriers assigned to
each of the position location signals is not equal.
[0055] In an embodiment where Q/M is an integer, the first position
location signal is an orthogonal frequency multiplexed signal that
can be chosen by initially randomly (or pseudorandomly) selecting a
first carrier from the set of Q orthogonal carriers. The selected
carrier is then removed from a list of available carriers. A second
carrier is randomly selected from the remaining Q-1 carriers. That
carrier is then also removed from the list of available carriers.
The process is continued in this manner until Q/M carriers are
selected and assigned to the first position location signal. Next
Q/M carriers are randomly selected for the second position location
signal from the remaining Q-Q/M carriers. This procedure is
continued until all M multiplexed position location signals are
assigned Q/M carriers. In this embodiment, the M subsets of
carriers assigned to each of the M multiplex signals are disjoint,
or mutually exclusive, from one another. The orthogonality of the
individual carriers ensures orthogonality of the position location
signals to one another, assuming perfect time and frequency
synchronization. This procedure for assigning frequencies to
position location signals may be repeated, with use of a new
randomized sequence, if it does not result in signal with desirable
properties, for example, autocorrelation properties with peak to
sidelobe ratio greater than some predetermined threshold.
[0056] In the example illustrated in FIG. 2, the total number of
carriers, Q, is equal to sixteen and the number of position
location signals, M, is four. Thus, each position location signal
includes a subset of Q/M=4 separate orthogonal carriers. A first
position location signal is assigned carriers numbered 1, 3, 7, and
12. A second position location signal is assigned carriers numbered
2, 6, 11, and 14. A third position location signal is assigned
carriers numbered 4, 9, 13, and 16. The fourth and final position
location signal of this example is assigned carriers numbered 5, 8,
10, and 15.
[0057] In another embodiment, the Q/M carriers for at least one of
the position location signals are uniformly spaced throughout the
overall signal bandwidth. A consequence of replacing random
frequency interlacing with a uniform interlacing is approximately
an M fold autocorrelation ambiguity. A uniform carrier spacing
results in an autocorrelation function repeating with a time
interval of T.sub.s/M, where T.sub.s is the basic OFDM symbol
period (without cyclic extension) and is the reciprocal of carrier
spacing. A uniform carrier spacing can result in reduced cross
correlation between sets with adjacent frequencies and reduce the
unambiguous delay range compared to an embodiment having random or
pseudo random carrier spacing. Nevertheless, in some situations
such a uniform spacing may be desirable for implementation reasons,
or required based upon system constraints.
[0058] The position location signal transmitted by a beacon is
generated using one or more of the subsets of carrier frequencies.
To generate the position location signal, signaling data is
assigned to each of the M carrier subsets, each of size Q/M
carriers. Typically, the signaling data takes the form of a
specified carrier phase, relative to other carrier phases at the
beginning of an OFDM symbol period. Thus, in this case a vector of
Q/M carrier phases is the signaling data associated with one OFDM
symbol of a given carrier subset. However, it is also possible to
utilize signaling data containing both carrier phase and amplitude
information, for example, quadrature amplitude modulation (QAM).
Normally the signal information assigned to a given carrier is held
constant over the OFDM symbol period, in order to maintain
orthogonality between differing carriers over the time period
T.sub.s.
[0059] In one embodiment, M random or pseudorandom binary sequences
are defined, and each of the sequences is assigned as a data vector
to one of the M carrier subsets. The binary sequences can be used
as a data source whose elements are used to assign the phases of
the various carriers. Each bit in the binary sequence can
correspond to a constant phase applied to each of the carriers in
the corresponding subset. In this embodiment, the phase of each
carrier remains fixed over an OFDM symbol period. It may be
advantageous to select the binary sequences from a special class or
sequences, such as maximal length PN sequences, Barker codes, Walsh
codes, or Gold Codes having a length truncated to or extended to
Q/M. Notice that the binary sequences described in this paragraph
are not used to directly modulate a signal versus time, but are
used to modulate a set of carriers versus frequency. Another way to
visualize this process is that the OFDM symbol is a linear
combination of carriers that belong to a specified subset. The
coefficients of the linear combination are the binary or higher
order sequences discussed above. For full generality then, i.e. to
produce phase and amplitude modulation, these sequences are complex
numbers.
[0060] Thus, there can be at least two levels of randomization for
each of the position location signals. A first level of
randomization includes a random frequency interleaving of M
signals, and a second level of randomization includes a random
assignment of phases (or amplitudes and phases) to the respective
Q/M frequencies of each of the M signals. In other embodiments, the
binary (or higher order) sequence data source may vary with time
(for example, may vary from one OFDM symbol to the next) and thus,
each of the carriers may be modulated with a time varying data
stream. Normally, however, that sequence data is still held fixed,
or nearly fixed, over each individual OFDM symbol period. Otherwise
orthogonality of the OFDM symbols will not be maintained.
[0061] Each of the M composite signals described in the above
embodiments can be concurrently transmitted from a corresponding
one of M distinct position location beacons as an OFDM symbol.
[0062] FIG. 3 is a functional block diagram of an embodiment of a
frequency interlaced orthogonal frequency position location beacon
120. The position location beacon 120 embodiment of FIG. 3 can be,
for example, any of the position location beacons 120a-120n in the
position location system 100 of FIG. 1.
[0063] The position location beacon 120 includes an orthogonal
signal generator 310 coupled to a modulator 320. The orthogonal
signal generator 310 can be configured to generate the subset of
frequencies used by the position location beacon 120. A modulation
data module 324 is also coupled to the modulator 320 and supplies
the binary (or other) sequence used to modulate the orthogonal
carriers. That is, the phase and/or amplitude of each carrier is
altered in accordance with an element of the sequence.
[0064] The modulated output of Q/M carriers from the modulator 320
is coupled to an input of an OFDM modulator 330 that can be
configured to generate the OFDM symbol corresponding to the
modulated carriers. The OFDM symbol output from the OFDM modulator
330 can be coupled to a Digital to Analog Converter (DAC) 340 where
a digital representation of the signal is converted to an analog
representation.
[0065] The output of the DAC 340 can be coupled to a transmitter
350 where the OFDM symbol is frequency converted to a broadcast
band. The output of the transmitter 350 can be coupled to an
antenna 360 for broadcast.
[0066] The orthogonal signal generator 310 can be configured to
generate all of the orthogonal carriers defined in the overall
signaling bandwidth. The orthogonal signal generator 310 may
generate, for example, all of the orthogonal carriers and pass only
those corresponding to the desired subset of frequencies and
filtering out the others. Alternatively, the orthogonal signal
generator 310 may be configured to generate a subset of orthogonal
frequencies including at least the subset of orthogonal frequencies
associated with the position location signal broadcast by the
particular position location beacon 120. The orthogonal signal
generator 310 can be configured to filter out any carriers that are
not part of the assigned frequency subset prior to outputting the
carriers to the modulator 320. The orthogonal signal generator 310
may be configured to alter the assigned frequency subset from one
OFDM symbol period to the next, or it may be configured to leave
the assignment unchanged.
[0067] As described earlier, the modulator 320 can be configured to
modulate each of the carriers in the subset with data supplied by
the modulation data module 324. In one embodiment, the modulation
data module 324 is configured to supply a pseudorandom sequence,
such as a Gold code with a length extended, or truncated, to Q/M,
to the modulator 320. The modulator 320 can be configured to
modulate a phase of each of the carriers in the subset of carriers
based on a corresponding bit value in the pseudorandom sequence.
For example, the modulator 320 can be configured to Binary Phase
Shift Key (BPSK) modulate the carriers based on the values of the
pseudorandom sequence. Alternatively it may be configured to
provide Quaternary Phase Shift Keying (QPSK), higher order phase
shift keying, quadrature amplitude modulation (QAM) or other forms
of modulation, based upon the values of a pseudorandom, or other
prescribed numerical sequence.
[0068] As a simple example, the modulator 320 can be configured to
modulate a carrier with a first phase, such as 0 degrees, in
response to a "1" in the pseudo random sequence. The modulator 320
can also be configured to modulate a carrier with a second phase,
for example 180 degrees, in response to a "0" in the pseudorandom
sequence. Therefore, if the modulation data module 324 supplies a
pseudorandom sequence of "0110" to the modulator 320 for the
embodiment where Q/M=4, the modulator phase may modulate the first
and fourth carriers to 180 degrees and phase modulates the second
and third carriers to 0 degrees. In effect, each of the carriers is
phase modulated with a constant value over one (or more) OFDM
symbol periods.
[0069] The finite number (M) of different frequency subsets may
result in the need to reuse the same frequency subset for more than
one beacon within the position location system. The system may
implement some manner of frequency subset reuse plan to alleviate
the effects of such duplication. The position location beacons
assigned the same frequency subset may also be assigned different
pseudorandom modulation codes in order to provide a receiver with
some ability to distinguish them.
[0070] The OFDM modulator 330 can be configured to convert the
phase (or phase and amplitude) modulated subset of frequencies to
an OFDM symbol. An embodiment of the OFDM modulator 330 can include
an Inverse Fast Fourier Transform (IFFT) module 334 configured to
convert the orthogonal frequencies to a time domain symbol. The
IFFT module 334 can be configured to perform the transformation on
the entire set of orthogonal frequencies, although only a subset of
frequencies contains information. If an IFFT module is used, it
typically produces a basic symbol of duration equal to the
reciprocal of the frequency line spacing. Often, the basic symbol
is extended to transmitted symbol duration by means of cyclic
extension, which could be considered to be part of the IFFT
processor 334. The output of the IFFT module 334 can be coupled to
a parallel to serial converter 338 to convert the parallel
generated time domain symbols from the IFFT module 334 to a serial
format. Other non-FFT implementations may be more cost effective,
particularly if the number of assigned frequencies is small. For
example the OFDM symbol may be constructed by modulating the
outputs of a series of digital oscillators. These oscillators may
be implemented by a number of means such as numerically controlled
oscillators.
[0071] The various modulation elements 310, 320, 330, 324 may be
timed with respect to one another and with respect to other events
(for example absolute time) by means of a system clock 370, which
in turn receives timing information from an external source. In
particular this external source may take the form of a signal from
a GPS receiver or from other terrestrial and/or space borne timing
sources. By this means the position location beacons may
synchronize the times of transmissions of their OFDM symbols to one
another, which in turn improves the orthogonality of their
transmissions relative to one another.
[0072] The operation of the position location beacon 120 can be
illustrated with a system example. As noted earlier, it may be
advantageous for the overall signaling bandwidth to be relatively
wideband to facilitate time resolution in the receiver.
Additionally, it may be advantageous to implement the position
location system within an existing communication system.
[0073] In one embodiment, the position location system can be
implemented within a television broadcast system. Television
broadcast systems are already widely supported and provide
extremely high transmit powers that span numerous relatively
wideband channels. The position location signal can be periodically
substituted for a normal television broadcast signal. Similarly the
position location signal may be periodically substituted for other
signals associated with wideband broadcast systems transmitting
video, audio or other information with high information
content.
AN EXAMPLE
[0074] In a typical wideband broadcast system embodiment, Q=4096
orthogonal carriers can be defined within a 5.5 MHz frequency span
that fits nicely with a typically allotted 6 MHz TV broadcast
channel. This allocation of carriers results in a carrier spacing
of approximately 1.343 kHz.
[0075] If M=8 different position location signals are configured,
each position location signal includes a subset having Q/M or 512
carriers. Each position location beacon 120 can be assigned one of
the eight carrier subsets.
[0076] The orthogonal signal generator 310 can be configured to
generate at least the 512 orthogonal carriers needed for a position
location signal. The 512 orthogonal carriers are then provided to
the input of the modulator 320. The modulation data module 324 is
configured to provide the modulator 320 a pseudorandom binary
sequence, such as a Gold code with a length extended to 512 bits if
required (full Gold code lengths are 2.sup.n-1, where n is an
integer, for example, for n=9 the length is 511). The modulator 320
can be configured to phase modulate each of the 512 carriers based
on the value of a corresponding bit in the length extended Gold
code. The OFDM symbol period should be at least 1/1.343 kHz=745
microseconds. Extending the period beyond this duration is, in
effect, a cyclic extension of the OFDM symbol since the composite
OFDM basic repeats in time after 0.745 microseconds, in this
example. Such cyclic extension allows for large differential delays
to be observed by a mobile station receiving signals from a
multiplicity of beacons; yet orthogonality may still be maintained
over the time interval of 745 microseconds.
[0077] The 512 phase modulated carriers output from the modulator
320 are coupled to an OFDM modulator 330. An IFFT module 334 within
the OFDM modulator 330 can perform a 4096 point IFFT operation.
Although the IFFT module 334 can be configured to perform a 4096
point IFFT, the 512 orthogonal carriers of the active subset may be
the only frequency bins having any nonzero signal component. The
corresponding 4096 bin output from the IFFT module 334 can be
coupled to a parallel to serial converter 338. The parallel to
serial converter 338 may include two parallel to serial converter
circuits because the IFFT typically produces complex data-that is,
in phase and quadrature data streams, which are sent to a pair of
DACs.
[0078] The serial outputs can be coupled to a pair of DACs 340 and
transmitter 350 which is coupled to an antenna 360, which may be
positioned on a broadcast tower that is itself positioned on a
geographic high point. The transmitter 350 can be configured to
periodically broadcast the position location signals. The
transmitter 350 itself typically includes a quadrature modulator
and additional upconversion and amplification circuitry to provide
the transmitted signal at an appropriate final RF frequency and
with requisite RF power.
[0079] The transmitter 350 can be configured to broadcast the
position location signal as an OFDM symbol that occurs once per
predefined repetition period. Periodic transmission allows the
position location signaling to be multiplexed with signals of
existing communication systems. The repetition rate can be, for
example, once per second or once per quarter of a second. Better
performance may be achieved, at a cost of reduced efficiency of the
underlying communication system, by sending the position location
symbols more often.
[0080] The position location beacon 120 can be configured to allow
for position location in the presence of large differential delays
between the signals from the various position location beacons and
the receiver. As indicated above, the position location beacon 120
can be configured to append a cyclic prefix to the data by
incorporating a cyclic extension approach in order to handle large
differential path delays. Various approaches can be used, such as
applying cyclic prefixes or suffixes to the data. For simplicity
the following discussion assumes use of cyclic prefixes.
[0081] In the above embodiment having 5.5 MHz bandwidth and 4096
carriers, the position location (basic) symbol is approximately 745
.mu.sec in length without the cyclic prefix, and can be increased
beyond this in length including the prefix. At the receiver a
portion of data of length 745 .mu.sec is retained and processed,
thus providing full sensitivity allowing for differential delays up
to the length of the cyclic prefix extension. Longer differential
delays may be processed, but there would be a loss in
sensitivity.
[0082] The position location ranging signals transmitted by a
position location beacon 120 may be configured for very long
delays, in order to accommodate reception of signals from beacons
over a wide geographical area. With sufficient transmit power, it
is possible to achieve a good probability of detection for beacon
signals transmitted at ranges in excess of 200 km, which
corresponds to a delay of about 667 microseconds. In order to
achieve this range with no loss in sensitivity, the cyclic prefix
would need to be 667 microseconds, thus extending the transmitted
symbol period to 745+667=1412 microseconds. If smaller ranges are
acceptable, then a shorter transmitted symbol period may be
allowable. Of course, when receiving and processing the OFDM
symbol, the receiver integrates the received signal over a 745
.mu.sec period.
[0083] In some situations, the position location signaling is time
shared with communication signaling. Then it is desirable that the
numerology of the two transmission types (communications and
position location) be commensurate. Typically, however, the range
requirements of the communication system are less. For example, in
the prior example, with 745 .mu.sec period per basic OFDM symbol
(not including cyclic prefix), assume that a cyclic prefix of 55
.mu.sec is used for communications purposes, thus producing a 800
.mu.sec transmitted symbol period. The 55 .mu.sec prefix may not be
long enough to accommodate the differential delays between beacons
for position location purposes, although it may be perfectly
adequate for communications purposes. This is due to the fact that
for communication purposes, the receiver need only communicate with
one transmitter, whereas for position location purposes the
receiver should be capable of concurrently receiving signals from
three or more transmitters (beacons). In this case one approach
would be to increase the cyclic prefix such that the transmissions
employed for position location equal two or more transmitted
symbols, for example 1600 .mu.sec in this example. This would imply
a cyclic prefix of duration 1600-745=855 .mu.sec, which would allow
for processing of signals transmitted from ranges up to 256 km.
However, it is easy to see that the unambiguous range corresponds
to delays of from 0 to 745 .mu.sec. Due to the repetition of the
signaling waveform every 745 .mu.sec, the receiver may not be able
to discriminate between a signal received at time delay d or d+745
.mu.sec. Normally, this ambiguity may be resolved by measuring the
received signal power level. An extra delay of 745 .mu.sec would
typically greatly weaken a signal's received power, thus permitting
ambiguity resolution via received power measurement.
[0084] In an alternative embodiment, where the transmitted symbol
period is restricted to 800 .mu.sec, the position location signal
may be constructed from a subset of 2048 carrier signals spaced
over the assumed 5.5 MHz passband; hence adjacent tones are spaced
by 2.69 kHz. This corresponds to a basic cycle period of 1/2.69
kHz=or 372 .mu.sec. In this case, since the assumed transmitted
symbols are restricted to 800 .mu.sec in duration, one sees that
the cyclic prefix is in effect of length 800-372=428 .mu.sec. This
provides a range of up to about 128 km. A delay ambiguity of 372
.mu.sec may result in this case.
[0085] In some embodiments, the subset of frequencies and pseudo
random sequence provided by the modulation data module 324 are not
varied. In these cases the OFDM symbol output from the OFDM
modulator 330 will also remain the same. In these cases, the
orthogonal signal generator 310, modulator 320, modulation data
module 324 and OFDM modulator 330 can be replaced by a module that
stores and periodically provides the same OFDM symbol associated
with the position location beacon 120.
[0086] FIG. 4 is a plot of normalized autocorrelation for a
position location signaling embodiment. The performance is shown
for a position location signal generated using Q=4096, M=8, and a
Doppler shift of 50 Hz. The RMS circular autocorrelation sidelobes
for each of the position location signals is down from the mainlobe
by a factor of (Q/M).sup.-0.5. For example, if Q=4096 and M=8, the
autocorrelation sidelobes are down by 1/sqrt(512), or about -27
dB.
[0087] FIG. 5 is a plot of cross correlation between two different
OFDM symbols with respect to frequency offset for a position
location signaling embodiment. The frequency offset is typically
due to Doppler shift associated with the motion of the receiver.
FIG. 5 shows curves plotted for two different position location
signaling configurations. A first configuration includes Q=4096 and
M=8 and a second configuration includes Q=2048 and M=8. With zero
Doppler, the circular cross correlation of the different
multiplexed position location signals is essentially zero. With
nonzero Doppler, the cross correlation performance is attributable
to a number of factors. The cross correlation rejection is partly
attributable to the minimum frequency separation of the signal
sets, the embedded pseudorandom binary code assigned to the
differing frequency components, and the fact that only a fraction
of frequencies of an interfering signal are typically within a
minimal distance from a frequency of a given position location
signal, where the minimal distance is the carrier spacing.
[0088] An approximate formula for the cross-correlation rejection
(in dB) is: 10.times.log.sub.10(.delta.f.times.T.sub.f)-10
log.sub.10(Q)+3, where .delta.f is the Doppler shift, or other
frequency offset, T.sub.f is the reciprocal of tone spacing (the
basic symbol length) and Q is as before the number of carrier
tones. Note that the cross correlation rejection does not depend
upon the number of multiplexed signals (M). This is contrast to the
autocorrelation rejection whose sidelobe structure depends upon the
number of channels. As seen in the plots of FIG. 5, the theory is
within 1 dB of the measured over the entire range. The difference
between measured and theory is probably dominated by errors in
approximation.
[0089] The effects of Doppler on the cross correlation performance
can be further reduced by generating position location signals
utilizing only half of the available carriers. The cross
correlation improvement is relatively small even though the lines
are spread apart by a factor of two. The spaced carrier embodiment
may be of interest because it can be configured to allow for large
differential time delays by including a cyclic prefix of greater
duration, while keeping the transmitted symbol duration the
same.
[0090] The nature of FIG. 5 suggests a method of discriminating
valid correlation peaks from spurious ones that might be caused by
large Doppler shifts. First, it should be noted that a valid peak
would have a correlation peak that varies as 20
log10(sinc(.delta.f.times.T.sub.f)). For most terrestrial cases of
interest this variation is negligible. For example, if T.sub.f=372
.mu.sec, the variation for differential Doppler shifts in the range
[-200 Hz, 200 Hz] is less than 0.1 dB. To determine if a received
signal is a valid correlation peak or if it is a cross correlation
peak, a receiver can vary the reference signal in frequency
increments of f.sub.i over a specified maximum range, say
[-200,+200 Hz]. If the correlation peak drops by greater than of
equal to some predetermined amount, say 3 dB, then reject the peak
as a cross correlation. For example, if f.sub.i were 50 Hz, then
one of the sets of increments would cause the spurious frequency
lines to be within 25 Hz of the nulls of the lines associated with
the test signal. From FIG. 5 the rejection at 25 Hz offset for the
Q=2048 case is approximately -71.4 dB. In effect then offsetting
the receiver reference can extend the Doppler rejection to around
-71.4+3=-68.4 dB, even for a relatively large Doppler range of
[-200,+200] Hz.
[0091] Another way to formulate the above test is simply to perform
the cross correlation operation for each of the test frequencies
and then choose at each lag, the smallest magnitude of cross
correlation, over the frequency range. The resulting set of data
can then be used to perform a detection operation.
[0092] The cost of performing cross correlation by incrementing the
frequency reference is a set of additional tests. In the above
example nine tests are used to cover a Doppler span of [-200, 200]
Hz. Often the processing burden is not too high if the requirement
to perform position location occurs infrequently. In this case
processing can be performed offline by a microprocessor or a
Digital Signal Processor (DSP). Furthermore, the processing can be
serialized over the set of frequencies and the set of candidate
signals, and hence storage can be a minor issue. The effects of
additive noise will not invalidate these tests because the noise
peak amplitudes typically vary little over these frequency ranges.
In principle, in addition to discriminating against false alarms,
this method can be used to "null out" the contribution of
interfering signals, and reveal the presence of a weak signal.
[0093] The cross correlation discrimination test should work well
because in the majority cases of interest there should be at most
one very strong signal that might cause detectable cross
correlations. The cross correlation issue is primarily associated
with the "near-far" type problem, which should typically be limited
to one emanation from one strong beacon and perhaps multiple weak
signals. For most cases of interest the interfering signal causing
the major cross-talk situation is extremely strong, for example 50
or 60 dB above any detection threshold. Typically, such a signal is
easily detectable.
[0094] Since the major interferer is likely to be very strong in
power, the precise Doppler frequency of such a signal may also be
measurable, by a number of methods, thus reducing the range over
which the above discrimination test may be performed. For example
following an initial detection, one can examine the signal
amplitude at Doppler frequencies displaced from this value, for
example at .+-.1/(4T.sub.f). These three amplitudes can then be
used in a quadratic interpolation process to accurately determine
the true Doppler. More optimal methods are also possible, at the
expense of more computation. It can be shown that the optimal
frequency estimator has an RMS value bounded below by the
Cramer-Rao bound as: 12 2 .times. .pi. .times. .times. T f .times.
1 2 .times. .times. SNR out .apprxeq. 0.39 T f .times. SNR out
##EQU1## where SNR.sub.out is =2 E/N.sub.0 is the output
signal-to-noise ratio (E is signal energy over T.sub.f and N.sub.0
is two-sided noise density), measured at the peak of a matched
filter's output.
[0095] A quadratic interpolation algorithm provides extremely good
performance. Using this process, the resulting Doppler estimates
can be generated and have RMS errors of 10.2 Hz and 1.1 Hz for the
60 dB and 40 dB SNR.sub.out situations respectively.
[0096] Hence, if the strong beacon signal has 60 dB output SNR, a
receiver can likely estimate its Doppler to an accuracy of around 1
Hz. This may eliminate the requirement to step through the Doppler
band, as indicated above. However, there are various effects that
may limit such an accurate estimate, including the presence of
smaller multipath reflections having differing Doppler shifts,
presence of other interfering signals, computational limitations,
and other factors.
[0097] FIGS. 6A-6C provide embodiments of receivers 600 that can be
used to receive the position location signals generated by the
position location beacon 120 of FIGS. 1 and 3. FIG. 6A is a
functional block diagram of an embodiment of a position location
receiver 600 configured to detect the broadcast position location
signals using a correlator 630. The position location receiver 600
can be implemented, for example, within a mobile station, such as
the mobile device 110 of FIG. 1.
[0098] The position location receiver 600 includes an antenna 602
configured to receive one or more signals in one or more frequency
bands. For example, the antenna 602 can be configured to receive
position location signals from a position location beacon 120 at a
first frequency band and a position location source in a second
communication system at a second frequency band, such as from the
base stations 130a-130b shown in FIG. 1.
[0099] The antenna 602 is coupled to an RF receiver 610 that can be
configured to amplify, filter, and frequency convert the received
wireless signal to, for example, a baseband signal. Because the
beacons may transmit the position location signals as a burst that
is time multiplexed with other communications, the RF receiver 610
can be configured to receive signals during a time allocated to
position location signals. The RF receiver 610, and succeeding
circuitry 630, 632 and 640, may be synchronized to a time reference
that is also used by the position location beacons. The output of
the RF receiver 610 can be coupled to an Analog to Digital
Converter (ADC) 620 that is configured to convert the signal to a
digital representation. The output of the ADC 620 can be coupled to
buffer memory or register bank 624. The memory 624 can be
configured to store the received position location signal for
further processing.
[0100] The receiver 600 can be configured to perform signal
detection, cross correlation rejection and Doppler processing in
the manner described above. The output of the memory 624 can be
coupled to a correlator 630. An OFDM symbol generator 632 can be
coupled to another input to the correlator 630. The OFDM symbol
generator 632 can be configured to generate each of the position
location symbols that are broadcast by the position location
beacons. Additionally, the OFDM symbol generator 632 can be
configured to generate frequency offset versions of the position
location signals to assist in cross correlation rejection and
Doppler determination. In the case where the carriers used in the
various position location signals are not varied and where the
carriers are not modulated with time varying data, the OFDM symbol
generator 632 may be configured to retrieve copies of previously
generated symbols and need not utilize computational circuitry to
generate the symbols for each correlation.
[0101] The correlator 630 can be configured to correlate the
received signal against the symbols generated by the OFDM symbol
generator 632. The correlator 630 can be configured to successively
correlate the input data against each candidate OFDM symbol or it
can perform the correlation operations simultaneously or otherwise
perform the correlation functions concurrently. Because the
position location beacons may burst the position location signals,
the correlator 630 need not perform the correlation functions real
time, and may perform the correlation functions during the period
of time in which the position location beacons are not broadcasting
position signaling. The memory serves the function to hold a burst
of signal energy that is used in this non realtime correlation
process.
[0102] The correlator 630 typically performs correlation operations
against the input signal for each of a series of assumed
time-of-arrivals, sometimes called "lags." The resulting series of
numbers is called a "sample cross-correlation function". Hence the
correlator 630 typically provides to the succeeding circuitry
sample cross-correlation functions for each candidate OFDM symbol
and perhaps for a multiplicity of assumed Doppler frequencies as
well. Often the inputs to the correlators 630 are two data streams,
an I and Q data stream, corresponding to in phase and quadrature
signal tributaries. In this case the correlators typically produce
sample cross-correlation functions that also contain I and Q
streams, or tributaries. One often considers these I and Q streams
as a single complex data stream. Often an envelope detection or
magnitude-squared operation is performed upon this complex data
stream, thus providing a single real data stream upon which
succeeding operations (such as signal detection) are performed.
[0103] The output of the correlator 630 can be coupled to a
position determination module 640. The position determination
module 640 can be configured to perform a portion of a position
location operation or may perform an entire position location
operation based at least in part on the output of the correlator
630. In one embodiment, the position determination module 640 can
be configured to determine the location of the receiver 600 by
performing trilateration to the originating position location
beacons. In other embodiments, the position determination module
640 can be configured to perform a portion of the position location
operation. For example, the position determination module 640 can
be configured to determine pseudoranges corresponding to the each
of the received position location signals. The position
determination module 640 can then communicate the pseudoranges to a
remote processor or server that is configured to determine the
location of the receiver 600. The position determination module 640
can, for example, transmit the pseudoranges to a server that is
part of a position location module, for example 140 from FIG. 1,
within a communication system separate from the communication
system that is used to generate the position location signals.
[0104] Another functional block diagram of an embodiment of a
receiver 600 is shown in FIG. 6B. The receiver 600 includes an
antenna 602 coupled to an RF receiver 610 as in the previous
embodiment. The output of the RF receiver 610 can be coupled to a
matched filter module 650. The output of the matched filter module
650 can be coupled to a peak detection module 652 that is
configured to determine whether the magnitude of the matched filter
output exceeds a predetermined threshold. The output of the peak
detection module 652 is coupled to a position determination module
640. The matched filter produces a sample cross-correlation
function that is mathematically equivalent to that provided by the
correlators 630; however, it performs this function by means of
filtering methods, as are well known in the art.
[0105] The matched filter module 650 can include one or more
matched filters configured to detect the position location signals.
In one embodiment, a plurality of matched filters is configured in
parallel, with each of the matched filters tuned to a specific
position location symbol or a Doppler shifter version of the
position location symbol. The impulse response of the matched
filter is the time-reversed conjugated version of the position
location symbol. In another embodiment, the matched filter module
650 includes at least one reconfigurable filter. The reconfigurable
filter is successively tuned to match one of the position location
symbols or Doppler shifted symbols. The received signal is then
provided to the reconfigurable filter. The matched filter module
650 can also include a combination of fixed filters and
reconfigurable filters.
[0106] In one embodiment discussed above, the received data
consists of a block of data of length 372 microseconds with an
unknown carrier phase transmitted each 0.86 seconds. The optimum
detection approach, assuming AWGN and ignoring multipath effects,
is to pass the signal through a matched filter module 650, compute
the magnitude of the signal, and look for a peak above a threshold
using the peak detection module 652. The threshold can be set based
upon a prescribed false alarm rate. If the receiver 600 is
configured to search over 8 different signal types, and Q/M=256
(Q=2048, M=8), then there are 2048 independent hypotheses. It may
be desirable to achieve a false alarm rate of no more than one
false alarm every hour (3600 seconds). This rate translates to a
false alarm rate per hypothesis on the order of
0.86/(2048.times.3600).apprxeq.10.sup.-7.
[0107] For a threshold k, the false alarm rate is approximately
exp(-T.sup.2/P.sub.N), where P.sub.N is the power of the noise in
the post processed bandwidth (the mean of the sum of the squares of
the I and Q variances following the matched filter). This implies
that the threshold should be set to about 12 dB. To further reduce
false alarms, the receiver 600 may require at least two successive
detections be achieved on successive transmissions and the maximum
expected Doppler.
[0108] FIG. 6C is another embodiment of a receiver 600. The
receiver 600 implements FFT techniques to detect the position
location symbols. In effect, the FFT method is an efficient means
toward implementing a matched filter. A signal processor can be
configured to perform an FFT method to select and appropriately
weight the various spectral components, according to the
hypothesized references. An inverse transform then produces the
matched filtered data. A feature of this approach is that a single
forward transform of the received data may be employed. Multiple
inverse transforms are required, one for each reference signal to
be tested. Additional inverse transforms may also be used to detect
Doppler shifted symbols.
[0109] The receiver 600 embodiment of FIG. 6C can include an
antenna 602 coupled to an RF receiver 610. The output of the RF
receiver 610 can be coupled to an ADC 620 that is configured to
convert the received signal to a digital representation. The output
of the ADC 620 can be coupled to an FFT module 660 that is
configured to perform a forward FFT on the received signal. The FFT
module 660 can be configured, for example, to perform the inverse
operation of that used in generating the position location symbol
at the beacon. For example, the FFT module 660 can be configured to
perform an FFT having a number of bins corresponding to the total
number of orthogonal carriers (Q).
[0110] The output of the FFT module 660 can be coupled to a
modulator (or multiplier) 670. A modulation data module 672 can
provide a binary (or other) sequence to the modulator 670. The
binary sequence is typically identical to the binary sequence used
to modulate the carriers in the position location signal beacons.
If the original position location beacon utilized a more
complicated sequence, then the modulator would typically provide a
sequence of numbers which are the complex conjugates of the
original complex sequence. Notice that this sequence of numbers is
indexed by carrier frequency number, not by a time index.
[0111] The modulator 670 can be synchronized with the modulation
data module 672 to apply a hypothesized reference to the results of
the forward FFT module 660. For example, the modulator 670 can be
configured to weight the output of the FFT module 660 to match
those of a particular position location symbol. The modulation data
module 672 can likewise be configured to provide the binary (or
other) sequence associated with the same position location symbol.
The modulator 670 and modulation data module 672 can be configured
to repeat the process for a hypothesis corresponding to each of the
position location symbols.
[0112] If phase modulation is used to modulate the data in the
position location beacons, the modulator 670 effectively operates
to remove the phase modulation on the subset of carriers to
generate a subset of unmodulated carriers. If binary phase shift
keying was originally used, then the phase removal is identical to
series of 180 degree phase reversals. If higher order phase
shifting was used, then the phase shifts applied to the various
carriers is simply the negative of the phase shifts applied to the
originating carriers. If phase and amplitude modulation (e.g. QAM)
is used, then the amplitude modulation is the same as in the
originating sequence, but the phase is again the negative of the
phase shift of the originating sequence.
[0113] The output of the modulator 670 can be coupled to an Inverse
FFT (IFFT) module 680. The IFFT module 680 can perform an IFFT
operation on the transformed received signals. The IFFT module 680
is typically of the same order as the FFT module 660. The output of
the IFFT module 680 can produce a correlation peak when the
hypotheses match the received position location signals. That is,
the presence of a peak indicates that the OFDM symbol hypothesis
and Doppler hypotheses are valid (or at least are approximately
correct). In addition the location of the peak relative to the
beginning of the data series indicates a relative time-of-arrival
of the signal from the corresponding beacon. Often the IFFT module
will perform a magnitude or magnitude-square operation upon the
inverse transformed data series, and it is this processed data that
is typically used in the peak search and detection operation.
[0114] The output of the peak detection module 690 can be coupled
to a position determination module 640. Thus, in an alternative to
performing direct time domain based correlation as in the
embodiments of FIGS. 6A and 6B, the receiver 600 can utilize
frequency domain methods to correlate the received signals to the
position location symbols using a combination of the FFT module
660, modulator 670, IFFT module 680 and peak detection module
690.
[0115] In order for the receiver 600 to support concurrent
processing of position location beacon signals from a range of
several kilometers through 120 km, the receiver 600 typically needs
around 55 dB of dynamic range. That is, signals should be
detectable and well above quantization noise effects. If the
receiver 600 requires at least 15 dB output SNR, the quantization
noise floor after the correlation processing should be about 70 dB
below the maximum correlation peak. In the above embodiment, the
integration time for the correlation process is set to 372 .mu.sec.
The effective processing gain is equal to the total number of
tones, or 2048, which corresponds to approximately 33.1 dB. This
implies that the signal-to-quantization noise ratio prior to the
correlation process be around 70-33.1=36.9 dB. This may correspond
to the desired ADC performance. The required performance depends
upon the sample rate of the ADC the type of ADC.
[0116] The worst case situation occurs if the ADC is a flash ADC
and if the filtering provided prior to the ADC is essentially a
brick wall type filter. In this case filtering between the ADC and
correlation operation will not improve the SNR. A q bit ADC has
approximately 2.sup.q levels. Depending upon the encoding, it may
be 2.sup.q-1 or 2.sup.q. Suppose the RMS input signal level is set
to 12 dB below the peak ADC output. This corresponds to an RMS of
1/ 4.times.(2.sup.q/ 2)=2.sup.q-3, assuming unity quantization step
sizes. The RMS quantization noise due to an ideal quantizer with
unity step size is 1/ {square root over (12)} and hence the RMS
signal to quantization noise is 20
log.sub.10(2.sup.q-3.times.sqrt(12)). This is shown in the
following table: TABLE-US-00001 RMS Signal to Quantization Ratio
No. Bits (dB) 6 28.9 7 34.9 8 40.9 9 46.9 10 52.9
[0117] From this table, in order to achieve 36.9 dB Signal to
Quantization noise the receiver 600 needs to have an 8 bit ADC. In
some cases, however, this requirement may be relaxed, as indicated
above, if, for example, the ADC sample rate is somewhat oversampled
and the quantization noise extends beyond the signal passband. Then
digital filtering can improve the effective signal strength. This
is particularly the case for sigma-delta converters in which much
of the quantization noise resides at the upper end of the passband
defined by half the sample rate.
[0118] The above requirement upon ADC quantization accuracy can
also be relaxed if a multiplicity of OFDM symbols are processed by
any of the methods of FIGS. 6A, 6B, or 6C and the results are
combined prior to performing the peak detection. The combination
may be done either prior to (coherent integration) or after
(incoherent integration) a magnitude (or magnitude-squared)
operation, depending upon whether continuous carrier phase may be
maintained between the OFDM symbols being processed. If coherent
integration is used, then the improvement in signal-to-quantization
noise is proportional to the number of OFDM symbols combined. For
example, if four symbols are combined the improvement is 6 dB,
hence reducing the ADC quantization accuracy by one bit. If
incoherent integration is used, then the improvement is less than
linear. For example, for a required 12 bit output SNR, if four
symbols are combined the improvement is approximately 5.1 dB
(assuming 10.sup.-7 false alarm rate). Hence, for this case about 5
symbols must be combined with incoherent integration in order to
reduce the ADC quantization accuracy by one bit.
[0119] As indicated earlier, the position location operation may be
time multiplexed with a communication function. In particular the
communication functionality may be one employing OFDM type
modulation. A communication receiver demodulating an OFDM stream
typically will perform a forward FFT as in 6C. However, it will
then normally process in the frequency domain each of the carriers
provided by the data out of 660 by demodulating the data present in
each of these carriers. This is in contrast to the position
location processing in which the data out of 660, following the
modulation function of 670, undergoes an inverse transform to
provide data in the time domain. Nevertheless the communication
receiver can share the processing stages of 610, 620, and 660 in
order to implement a very efficient combined communications and
position location system.
[0120] FIG. 7 is a flowchart of an embodiment of a method 700 of
generating position location signals. The method 700 can be
performed by the position location beacons 120a-120n of FIG. 1 and
FIG. 3.
[0121] The position location beacon processing can begin at block
710, which generates frequency interlaced OFDM position location
signals. The position location beacon can generate one or more
frequency interlaced OFDM signals, although typically, the position
location beacon is assigned only one of the signals.
[0122] The position location beacon proceeds to block 720 and
applies redundancy to the generated signal. As described above in
one of the embodiments, the position location beacon may generate a
(basic) symbol that is 372 .mu.sec in length but may repeat the
signal to generate a signal of 744 .mu.sec in length. This acts as
a cyclic prefix (or suffix) and permits a receiver to process
received signals at long ranges from the beacon without sensitivity
loss due to crosstalk from more nearby beacons. Of course much
shorter or longer cyclic prefixes may be used depending upon the
set of geographic ranges that need to be accommodated.
[0123] After applying the cyclic prefix, the position location
beacon can proceed to block 730 and synchronize the position
location signal to a time reference. As discussed earlier, the
position location beacons should be synchronized to minimize
position errors attributable to timing errors. The position
location beacon may be synchronized to an external timing
reference, such as GPS time or some other time reference. The
position location beacon timing should be fixed relative to the
other position location beacons and should be accurate to within
100 nsec and preferably accurate to within 50 nsec.
[0124] Once the position location beacon is synchronized to a time
reference, the position location beacon can proceed to block 740
and transmit the position location signal at a predetermined time
relative to the time reference. The position location beacon can
then return to block 710 of the method 700 to repeat the
process.
[0125] In some cases, the synchronization to the time reference
operation of block 730 is done concurrently with the operations of
710 and 720, particular if the implementation of 710 and 720 is
done in realtime with custom hardware. If instead, 710 and 720 are
done in software, for example by pre-computing a set of data
samples for later transmission, then the synchronization of this
data to a time reference would occur when such a transmission is
required, and the location of the synchronization function 730 is
appropriate.
[0126] FIG. 8 is a flowchart of an embodiment of the method 710 for
generating frequency interlaced OFDM position location signals. The
method 710 can form a part of the signaling flowchart of FIG. 7,
and can be performed in a position location beacon, such as the one
shown in FIG. 3.
[0127] The method 710 begins at block 810 where the position
location beacon generates a plurality of orthogonal carriers, Q,
that substantially span a channel bandwidth. As noted earlier, a
wider channel bandwidth facilitates a sharp correlation peak in a
receiver and allows for increased time resolution. In one
embodiment, the channel bandwidth can be approximately 5.5 MHz
wide. Embodiments discussed above have included Q=4096 and Q=2048,
although the number of orthogonal carriers is not limited to a
power of two.
[0128] After generating the Q orthogonal carriers, the position
location beacon proceeds to block 820 and selects a subset (Q/M) of
the orthogonal carriers. In one embodiment, the carriers in the
subset are uniformly spaced throughout the channel bandwidth. In
another embodiment, the carriers in the subset are randomly or
pseudo randomly spaced. Normally, the number of carriers in each
subset is chosen to be identical (that is, Q/M for M subsets).
However, it is possible, and in some cases desirable, that the
number of carriers in the M subsets may differ. In this case the
number of carriers may be more or less than Q/M but each of the
subsets should be disjoint in order to maintain orthogonality, and
the sum of the number of carriers in all subsets should be Q or
nearly Q. In some cases some of the carriers, particularly at the
lower and/or upper ends of the band, may be left unused in order to
provide a guard band that minimizes interference with adjacent
signals. In some cases some of the carriers may be utilized for
other purposes, such as synchronization, and hence may be
unavailable for position location purposes.
[0129] It may be advantageous for each of the subsets to have
carriers that are mutually exclusive of the other subsets. One of
the frequency subsets can be assigned to each of the position
location beacons in the position location system. Where there are
more beacons than distinct subsets of carriers, the system may
implement a reuse plan that minimizes the potential interference
from position location beacons having like frequency subset
assignments. In particular it may be advantageous for a beacon
numbered M+1 to utilize a different subset of frequencies than any
of the M distinct subsets mentioned above. Then, while the OFDM
symbol from this additional beacon will have some degree of
correlation with an OFDM symbol corresponding to one or more of the
M subsets, this correlation may be low.
[0130] After the position location beacon selects or is otherwise
assigned a subset of carriers, the position location beacon
proceeds to block 830 and modulates the subset of carriers. In one
embodiment, a pseudo random code, such as a Gold code of length Q/M
is used as a data sequence to BPSK (or otherwise, for example, QAM)
modulate the carriers. That is, the each carrier is modified in
phase and/or amplitude in accordance with an element of such a data
sequence. The pseudo random modulation data can be fixed or may
vary over time. Typically, the modulation data sequence assigned to
different subsets are chosen to be different. The different data
sequences are typically chosen to have good cross correlation
properties.
[0131] After modulating the subset of carriers, the position
location beacon proceeds to block 840. In block 840 the position
location module generates the frequency interlaced OFDM symbol
corresponding to the interlaced and modulated carriers. For
example, the position location beacon can generate the OFDM symbol
using a IFFT module and parallel to serial converter.
[0132] In the above description of FIG. 8, it should be noted that
the operations 810 to 830 may be viewed as simply constructing a
series of Q complex numbers, one number per each carrier frequency.
In this case, the operation 810 is simply the construction of an
array of Q numbers, each of which may be initialized to value 0.
The operation of 820 is the selection of the indices of the array
corresponding to a carrier subset and that of 830 is the assignment
of a phase and amplitude, or complex number, to each of the
elements of the array. This array may then be appended with
additional zero-valued samples at its beginning or end, in order to
create an array with a desired length (e.g. 2048, 4096 or another
power of two), and then the array may be operated upon with an
inverse FFT (or mathematically equivalent) operation to produce the
OFDM symbol as in 840.
[0133] FIG. 9 is a flowchart of an embodiment of a method 900 of
position location using frequency interlaced OFDM symbols. The
method 900 can be performed, for example, in the receiver 600
embodiments shown in FIGS. 6A-6C.
[0134] The method 900 begins at block 910 where the receiver
receives the position location signals. The receiver may be
synchronized to the same time reference used to synchronize the
beacons. Therefore, the receiver needs only tune to and monitor for
position location signals during a predetermined period of time.
The duration and duty cycle of the position location signals may be
only a fraction of the time that the receiver is active in order to
minimize the energy expended and processing required to support
position location. This is particularly the case if the position
location operation is time multiplexed with other operations, such
as communication signal processing.
[0135] After receiving the position location signals, the receiver
proceeds to block 920 and determines the received position location
symbols present in the received position location signals. The
determination can be a preliminary determination because some of
the symbols may be rejected as cross correlation products.
[0136] In 920 the receiver can, for example, correlate the received
signals to one or more reference symbols stored or generated within
the receiver. In another embodiment, the receiver may pass the
received signals through one or more matched filters corresponding
to the position location symbols. In still another embodiment, the
receiver may transform the signals to the frequency domain in an
FFT, remove the modulation on the carriers, and transform the
signals in an IFFT in order to determine the presence of particular
position location symbols. As indicated previously, this FFT
approach is particularly efficient if forward FFT processing is
utilized as part of a combined communication system/position
location system. Of course, the receiver may use some other
embodiment for determining the presence of position location
symbols.
[0137] After determining the presence of position location symbols,
the receiver proceeds to block 930 and can determine a Doppler
shift of the received signals using, for example, a quadratic
interpolation that may be further corrected by applying a
polynomial correction to the Doppler estimate. After determining
the Doppler shift, the receiver can proceed to block 940 and reject
the cross correlation products.
[0138] In one embodiment, the receiver shifts a frequency reference
by a predetermined amount above and below a nominal frequency. The
receiver can repeat the symbol determination for the frequency
shifted reference. The receiver can reject symbols as cross
correlation products for which a correlation peak drops by some
predetermined amount, for example 3 dB or more.
[0139] After rejecting the symbols that are determined to be cross
correlation products, the receiver can proceed to block 950 and can
determine valid received frequency interlaced OFDM symbols
corresponding to the position location beacons. These are basically
the symbols determined via operation 920 minus those rejected in
940. In some cases some additional rejections are performed. For
example, as indicated previously, there may be a time ambiguity
present due to the use of a long cyclic prefix. This may result in
an OFDM symbol appearing to be received at two or more different
times (spaced by the transmitted symbol period). Criteria such as
received power level may be used to perform such additional
rejections.
[0140] The receiver can then proceed to block 960 and determine a
time of arrival, or pseudorange, corresponding to each of the
received OFDM symbols. The receiver may have knowledge of the time
at which the symbols were transmitted. The receiver can determine a
signal delay, with a bias due to any error present in its local
clock, and thus determine a pseudorange to each of the transmitting
beacons.
[0141] The receiver can then proceed to block 970 and determine its
location, based at least ion part on the pseudoranges. The receiver
may independently determine its position, or the receiver may
transmit the pseudoranges to a position location module that
determines the location of the receiver.
[0142] In one embodiment, the receiver receives geographical and
other location information for the position location beacons over
an overhead data channel. The receiver is then able to determine
its position based on the beacon positions and corresponding
pseudoranges. In another embodiment, the receiver transmits the
pseudoranges to a position location module, such as a position
location server that is part of a cellular telephone position
location system. The receiver may transmit the information using a
wireless transmitter that is part of a module device that houses
the receiver.
[0143] Although a particular sequence of steps is shown in the
flowcharts of FIGS. 7-9, the methods are not limited to the steps
or sequence of steps shown in FIGS. 7-9. Additional steps or
processes may be added to the methods and the additional steps or
processes may be added between existing process steps. Moreover,
some steps or process flows may be omitted from the method. For
example, the method 700 shown in FIG. 7 may omit redundancy. As
another example, the method 900 of FIG. 9 may omit the cross
correlation rejection step.
[0144] A position location system, position location signaling,
position location beacon and receiver have been disclosed.
Additionally, methods for position location have been disclosed. A
frequency interlaced OFDM position location signal can be generated
from a set of orthogonal carriers spanning substantially a channel
bandwidth. The position location signals can be generated by
selecting a subset of the carriers. Each of the carriers in the
subset of carriers can be modulated according to an element of a
modulation data sequence. The modulation data sequence can be a
pseudo random sequence such as a Gold code sequence and the
carriers can be BPSK modulated with the data, or modulated with
higher order modulation, such as QAM. In the latter case the
elements of the modulating data sequence would have a higher order
quantization than binary. For example the modulating sequence may
contain a set of elements that are quantized to 3, 4 or more bits.
The modulated subset of carriers can then be transformed to an OFDM
symbol that is typically periodically transmitted. Multiple
position location beacons can be synchronized to transmit the
frequency interlaced OFDM symbols at substantially the same time.
Improved interference rejection results from having the frequency
subsets assigned to different beacons disjoint and designing the
modulation data sequences of different beacons to have good cross
correlation properties.
[0145] A receiver can receive the frequency interlaced OFDM symbols
from one or more position location beacons and can determine which
symbols were received. The receiver can then determine a
pseudorange based in part on the received symbol. The receiver can
then determine a location based on the pseudoranges.
[0146] A number of modifications are possible. For example, as
indicated previously, in some cases different beacons may utilize
different subsets of carriers, but the subsets may have some of the
carriers in common, rather than being totally disjoint. This will
cause some correlation to be present between the OFDM symbols
provided by the different beacons, but this correlation may be
minimized if the number of carriers in common is small.
[0147] The prior discussion focused upon situations in which the
multiple position location beacons transmitted information in a
time synchronized manner. In an alternative embodiment, the beacons
need not be time synchronized, as long as the times of transmission
of signals from the beacons may be ascertained. This may be done by
the use of mobile or fixed monitoring equipment that can time tag
the transmissions of signals from these beacons. For example,
cellular telephone equipment (mobile or fixed) containing GPS
receivers typically can determine accurately the time of day at
such receivers, and hence, if in proximity to beacons, can provide
time tagging of such transmissions. The relative timing of the
beacon transmissions may then be sent to a receiver trilaterating
from the beacon transmissions or to a server acting in conjunction
with this receiver that participates in position location. If the
beacons are not time synchronized then there is the potential that
some of the orthogonality between their emissions of symbols will
be lost. However, this problem may be avoided if the transmitted
symbols are repeated a number of times, thus obviating the
necessity for precise synchronization of such transmissions.
[0148] The above description is provided to enable any person
skilled in the art to make or use the invention. Various
modifications to the embodiments disclosed will be readily apparent
to those skilled in the art, and the generic principles defined
herein may be applied to other embodiments. Thus, the disclosure is
not intended to be limited to the embodiments shown herein.
* * * * *